README.md

README

Version: "v0.1.0"

Author: "Zijun Zhang"

Date: "2.24.2019"

Table of Contents

1. Installation
2. Testing
3. Using Darts BHT

Installation

Install via Anaconda

The recommended way to install Darts_BHT is through Anaconda. You can also create a new environment for Darts, because currently DARTS works in Python 2.7.

conda create -n darts python=2.7  # optional
source activate darts
conda install -c darts-comp-bio darts_bht

This will allow conda to do all the heavy-lifting and most often the easiest way to get things spinning.

Compile the source Rcpp package

Alternatively, to install Darts_BHT R package, you can browse to the BHT folder and type the following command in terminal:

cd Darts_BHT/
R CMD build Darts
R CMD INSTALL Darts_{versionNumber}.tar.gz

This will compile a few Rcpp files and prepare for loading. If finished successfully, you will see the following message

** building package indices
** testing if installed package can be loaded
* DONE (Darts)

Testing

Testing the Python wrapper

You can download the testing_data from the Github repo. If you installed through Anaconda, then type the following command and should see the corresponding help message as below:

> Darts_BHT -h
usage: Darts_BHT [-h] [--version] {rmats_count,bayes_infer} ...

Darts_BHT -- DARTS - Deep-learning Augmented RNA-seq analysis of Transcript Splicing

positional arguments:
  {rmats_count,bayes_infer}
    rmats_count         Darts_BHT rmats_count: run rMATS-turbo to count
                        junction reads in BAM files
    bayes_infer         Dart_BHT bayes_infer: perform Bayesian hypothesis
                        testing inference

optional arguments:
  -h, --help            show this help message and exit
  --version             show program's version number and exit

For command line options of each sub-command, type: Darts_BHT COMMAND -h

You can test run the installed Darts_BHT by:

Darts_BHT bayes_infer --darts-count test_data/test_norep_data.txt --od test_data/

Testing the R package

Now let's do a few quick checks to make sure the installed Darts_BHT R package works. There are a few built-in testing functions in Darts package. Open a new R interactive session, and run following code:

library(Darts)
set.seed(101)
t1=test_sim_darts(covg=200)
print(t1)

If you are not familiar with the RNA-seq splicing anlaysis, this is also a good time for you to get to know what's going on. See Here for a quick tutorial.

Also a quick test on replicate model; we will dive deeper into it in the Benchmarking section:

t2=test_sim_rdarts(n=10, outlier=F)
print(t2$eval_df)

Using Darts BHT

Now let's use a more wrapped-up script to call Darts BHT. In the simplest case, you only need to give Darts_BHT the input file -i and output folder -o. The data files are in the same format as rMATS, and can be generated by running rMATS w/ or w/o turning on the stat part.

For more concrete examples, please refer to the online doc Get Started at: here